• Journal of Gastric Cancer
• /
• 제6권3호
• /
• pp.167-172
• /
• 2006

목적: 조기위암 치료에 있어서 복강경 위암 수술이 새로운 패러다임으로 정착하고 있다. 기존에 시행하고 있던 개복에 의한 위절제술에 익숙한 경험 많은 외과의사들은 복강경 위암 수술은 시간이 많이 소요되며 기술적으로 습득하는데 많은 제약이 있음을 느끼며 복강경 수술에 소극적 자세를 취하는 경향이 있다. 이에 저자 등은 개복 위절제술에 경험이 풍부한 외과의사에 의한 복강경 위절제술의 learning curve를 결정하고, 이에 영향을 미치는 요인들을 분석하였다. 대상 및 방법: 2005년 4월부터 2006년 3월까지 한 명의 술자에 의하여 수술전 조기 위암(cT1N0)으로 진단 받고 복강경 보조하 원위부 위절제술(LADG) 및 $D1+{\beta}$ 림프절 곽청을 시행 받은 62명을 대상으로 하였으며 복강경 위절제술 시행을 위하여 전문팀을 구성하였다. 대상 환자를 6명씩 한 그룹으로 총 10그룹(마지막 그룹은 8명으로 구성함)으로 나누어 각 그룹의 평균 수술시간을 비교하여 learning curve 극복 전후의 나이, 성별, 수술 후 합병증, 절개창의 길이, 수혈유무, 적출된 림프절 개수, 수술 전 후 혈색소 변화 등을 분석하였다. 결과: 평균 수술 시간을 분석한 결과 여섯 번째 그룹 (31st case)부터 수술 시간의 Plateau를 보였다. 이에 저자들은 learning curve 극복시점을 30th case (7개월)로 간주하였으며 극복 전후 집단 간의 평균 수술시간을 분석하였을 때 각각 $239.0{\pm}69.7$분과 $170.0{\pm}32.6$분으로 유의하게 나타났다(P<0.05), 양 군 간 평균나이, 성별, BMI, 수술 전후의 혈색소 수치변화, 적출된 림프절 개수 등은 유의한 차이를 보이지 않았다. 또한 절개창의 길이, 수혈유무, 수술 후 합병증 유무도 양 군 간에 통계적으로 유의하지 않았다. 결론: 저자들의 LADG learning curve 극복은 30예(7개월)로 다른 보고보다 일찍 도달할 수 있었다. 이는 개복 수술의 풍부한 경험, 전문 수술팀 구성, 그리고 단기간 집중적인 시술에 의한 것으로 생각된다.

• 대한인간공학회지
• /
• 제11권2호
• /
• pp.47-56
• /
• 1992

In this paper, a mathematical model of learning curve is proposed to study the fi- nger's reaction time. The model is a logarithmic linear type which represents a lear- ning curve appropriately, and parameters are estimated by the linear. The learning coefficient and percentage of a reaction time can easily computed in the mathematical model. This quantitative approach provieds an important information to be used fot the working capqbility qualification of re-employment as well as the adaptability estimation of aged workers.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권13호
• /
• pp.5395-5399
• /
• 2014

Purpose: This study was designed to evaluate the outcomes of laparoscopic colorectal resection in a period of learning curve completed by surgeons with different experience and aptitudes with a view to making clear whether seniors had a better learning curve compared with juniors. Methods: From May 2010 to August 2012, the first twenty patients underwent laparoscopic colorectal resection completed by each surgeon were selected for analysis retrospectively. A total of 240 patients treated by 5 seniors and 7 juniors were divided into the senior group (n=100) and the junior group (n=140). The short-term outcomes of laparoscopic surgery of the two groups were compared. Results: The mean numbers of lymph nodes harvested were $21.2{\pm}11.0$ in the senior group and $17.3{\pm}11.5$ in the junior group (p=0.010); The mean operative times were $187.9{\pm}60.0min$ as compared to $231.3{\pm}55.7min$ (p=0.006), and blood loss values were $177.0{\pm}100.7ml$ and $234.0{\pm}185ml$, respectively (p=0.001); Conversion rate in the senior group was obviously lower than in the junior group (10.0% vs 20.7%, p=0.027) and the mean time to passing of first flatus were $3.3{\pm}0.9$ and $3.8{\pm}0.9$ days (p=0.001). For low rectal cancer, the sphincter preserving rates were 68.7% and 35.3% (p=0.027). Conclusions: Seniors could perform laparoscopic colorectal resection with relatively better oncological outcomes and quicker recovery, and seniors could master the laparoscopic skill more easily and quickly. Seniors had a better learning curve for laparoscopic colorectal cancer resection compared to juniors.

• 한국통신학회논문지
• /
• 제39B권8호
• /
• pp.518-527
• /
• 2014

본 연구는 유선통신 서비스 상품별 고객 불만(Voice of Customer) 감소에 따른 학습곡선을 추정 하고자 한다. 학습곡선모형 중 가장 일반적인 지수감소모형(Exponential decay model)을 사용하여 시간에 따라 고객 불만(VOC)이 감소하는지를 검증하였다. 그리고 통신사들의 서비스 상품의 인력투입, 소프트웨어 적용, 투자 등의 노력에 따른 고객 불만(VOC) 변화효과를 추가로 검증하였다. 서비스 상품별 실증 분석의 결과는 다음과 같다. 첫째, 학습곡선대로 시간에 따라 고객 불만(VOC)이 감소하였다. 둘째, 초고속 인터넷, 전화, IPTV 등은 인력투입, Network 장애, 계절요인으로 인해 고객 불만(VOC)을 증가 시키거나 감소 시켰다. 셋째, 서비스 상품별 다양한 변수는 고객의 체감 품질을 높이고 있지만, 오히려 지속적으로 감소하지 않는 서비스 패러독스(Service Paradox)현상이 발생하는 것을 알 수 있었다.

• 에너지공학
• /
• 제15권4호
• /
• pp.209-228
• /
• 2006

학습효과는 에너지기술의 전개와 관련된 메커니즘을 규명하려는 이론 중 하나이다. 본 논문에서는 학습 효과에 대한 이론적 고찰을 함으로써 아직까지는 국내에서 일천한 관련 이론에 대한 기반을 제공하고자 한다. 이를 위하여 학습곡선과 관련된 국내외 선행연구사례, 제반이론, 적용방법 및 정책 응용에 관하여 살펴본다. 또한 에너지기술의 학습과 비용절감 요인에 대하여 살펴봄으로써 학습곡선의 메커니즘을 파악한다. 마지막으로 각 장별 내용을 바탕으로 결론을 도출한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권9호
• /
• pp.2626-2641
• /
• 2024

Images taken in low lamination condition suffer from low contrast and loss of information. Low lumination image enhancement algorithms are required to improve the quality and broaden the applications of such images. In this study, we proposed a new Low lumination image enhancement architecture consisting of a transformer-based curve learning and an encoder-decoder-based texture enhancer. Considering the high effectiveness of curve matching, we constructed a transformer-based network to estimate the learnable curve for pixel mapping. Curve estimation requires global relationships that can be extracted through the transformer framework. To further improve the texture detail, we introduced an encoder-decoder network to extract local features and suppress the noise. Experiments on LOL and SID datasets showed that the proposed method not only has competitive performance compared to state-of-the-art techniques but also has great efficiency.

• International journal of advanced smart convergence
• /
• 제10권3호
• /
• pp.51-58
• /
• 2021

In this paper, we propose a deep learning classifier for estimating the number of layers in the Earth's structure. When installing a grounding system, knowledge of the subsurface in the area is absolutely necessary. The subsurface structure can be modeled by the earth parameters. Knowing the exact number of layers can significantly reduce the amount of computation to estimate these parameters. The classifier consists of a feedforward neural network. Apparent resistivity curves were used to train the deep learning classifier. The apparent resistivity at 20 equally spaced log points in each curve are used as the features for the input of the deep learning classifier. Apparent resistivity curve data sets are collected either by theoretical calculations or by Wenner's measurement method. Deep learning classifiers are coded by Keras, an open source neural network library written in Python. This model has been shown to converge with close to 100% accuracy.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권10호
• /
• pp.5159-5178
• /
• 2018

Network Intrusion detection is a rapidly growing field of information security due to its importance for modern IT infrastructure. Many supervised and unsupervised learning techniques have been devised by researchers from discipline of machine learning and data mining to achieve reliable detection of anomalies. In this paper, a deep convolutional neural network (DCNN) based intrusion detection system (IDS) is proposed, implemented and analyzed. Deep CNN core of proposed IDS is fine-tuned using Randomized search over configuration space. Proposed system is trained and tested on NSLKDD training and testing datasets using GPU. Performance comparisons of proposed DCNN model are provided with other classifiers using well-known metrics including Receiver operating characteristics (RoC) curve, Area under RoC curve (AuC), accuracy, precision-recall curve and mean average precision (mAP). The experimental results of proposed DCNN based IDS shows promising results for real world application in anomaly detection systems.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제5권3호
• /
• pp.263-267
• /
• 2005

Building classifiers for financial real-world classification problems is often plagued by severely overlapping and highly skewed class distribution. New performance measures such as receiver operating characteristic (ROC) curve and area under ROC curve (AUC) have been recently introduced in evaluating and building classifiers for those kind of problems. They are, however, in-effective to evaluation of classifier's discrimination performance in a particular class of the classification problems that interests lie in only a local operating range of the classifier, In this paper, a new method is proposed that enables us to directly improve classifier's discrimination performance at a desired local operating range by defining and optimizing a partial area under ROC curve or domain-specific curve, which is difficult to achieve with conventional classification accuracy based learning methods. The effectiveness of the proposed approach is demonstrated in terms of fraud detection capability in a real-world fraud detection problem compared with the MSE-based approach.

• Journal of Gastric Cancer
• /
• 제18권1호
• /
• pp.90-98
• /
• 2018

Purpose: We evaluated the learning curve and short-term surgical outcomes of robot-assisted distal gastrectomy (RADG) performed by a single surgeon experienced in open, but not laparoscopic, gastrectomy. We aimed to verify the feasibility of performing RADG without extensive laparoscopic experience. Materials and Methods: Between July 2012 and December 2016, 60 RADG procedures were performed by a single surgeon using the da $Vinci^{(R)}$ Surgical System (Intuitive Surgical). Patient characteristics, the length of the learning curve, surgical parameters, and short-term postoperative outcomes were analyzed and compared before and after the learning curve had been overcome. Results: The duration of surgery rapidly decreased from the first to the fourth case; after 25 procedures, the duration of surgery was stabilized, suggesting that the learning curve had been overcome. Cases were divided into 2 groups: 25 cases before the learning curve had been overcome (early cases) and 35 later cases. The mean duration of surgery was 420.8 minutes for the initial cases and 281.7 minutes for the later cases (P<0.001). The console time was significantly shorter during the later cases (168.6 minutes) than during the early cases (247.1 minutes) (P<0.001). Although the volume of blood loss during surgery declined over time, there was no significant difference between the early and later cases. No other postoperative outcomes differed between the 2 groups. Pathology reports revealed the presence of mucosal invasion in 58 patients and submucosal invasion in 2 patients. Conclusions: RADG can be performed safely with acceptable surgical outcomes by experts in open gastrectomy.